Contact Mics, Geophones, and Accelerometers, Oh My!

Vibration is vibration. Air molecules set in motion by a gentle whisper, stone and concrete trembling under the weight of a 2 ton shipping truck traveling at the speed of a bird in freefall. “بحل بحل” (Bhel Bhel), as a Moroccan friend of mine would say. “Same same”. Vibrations in solid matter, however, are more often felt than heard in our everyday experience. Assembled here are a number of experiments in bringing this physical vibration into the auditory realm.

Piezos

The simplest and most widely-used tool for converting material excitation into sound is a piezoelectric transducer. These can be found in a variety of form factors including discs, bars, and cylinders and can usually be obtained at prices ranging from inexpensive to positively dirt-cheap, making them a great first project for the DIY audio enthusiast. These transducers take advantage of a unique property of _____ crystalline structures, which when compressed by pressure waves (vibrations in a material to which the transducer is physically attached) generate a minute corresponding current. By connecting the output of the transducer to a speaker (usually through a preamp circuit of some kind), it becomes possible to experience that fluctuating current as sound. This process can be as simple as soldering the two materials of, say, a piezo disc to the ground and hot leads of an audio cable and plugging the other end of that cable into any audio recorder or amplifier.

*A note about soldering piezo discs: the strength of the piezoelectric effect of these elements can be reduced by exposure to extreme heat, so it is important not to let the soldering iron linger on the element too long. I find a medium temperature setting (~350C) and a quick touch with a fine flux-core solder to be reasonably effective.

These elements can be used in mono, stereo, and multi-channel configurations, or several can be chained in series.

Attachment Methods

In order for these piezo elements to be useful, they must be physically attached to the object one wishes to listen to. Some amount of pressure can help ensure full contact between the two and improve the transmission of low-frequency vibrations, but simply holding the piezo onto a surface is likely to introduce handling noise and at higher gain settings one’s own blood flow can introduce unwanted low-frequency information. Below are a few attachment methods I have tried so far:

Poster Adhesives

Poster adhesives like Blue-Tak can be great for temporarily sticking contact microphones to a wide variety of surfaces. This connection can be quite firm, sometimes so much so that it becomes difficult to pry the disc off! It can leave some residue on either surface (Plasti-Dip users beware) and does reduce the amount of surface contact between the element and the surface, but can be a great run-and-gun option for situations in which other attachment methods are impractical.

Clamps

For attaching contact mics to thin materials (tables, I-beams, paper, etc), simple spring clamps can be a quick and effective attachment option. The only obvious downside is that one is restricted to material thicknesses that fit within the range of the clamp used, and it is possible to break fragile piezo elements by exerting excessive force (ask me how I know). Clamps are also bulky and take up more space in transport, but I have found the Irwin Quick-Clamps to be a regular addition to my field recording kit.

Magnets

Magnets glued to the back of a contact mic can be an excellent option for quick mounting to metallic surfaces like fences, poles, and structural pylons, and as a bonus don’t get in the way of other attachment methods for non-ferrous materials. My current every-day-carry contact microphones feature small magnets glued to the back under a layer of Plasti-Dip, and I find them more useful than I ever expected.

DIY Piezo Preamp

Piezo elements output signal at a very high impedance, giving them a reputation for sounding thin and “honky.” This can be remedied with the use of an impedance-matching preamp. I built a simple one using this PCB by Jules Ryckebusch (the same one used in the Gladys Hydrophones) mounted inside an old Behringer DI box I had lying around. This preamp requires Phantom Power to function, but drastically improves the frequency response and sensitivity of any piezo contact mic.

Geophones

Geophones are sensors commonly used for seismic monitoring and oil prospecting. As such, they are highly sensitive to frequencies from ~200-500 Hz down into the infrasound range and are of course uniquely suited for recording vibrations that travel through the ground. There are a few commercial geophones for audio recording available (the LOM units are quite popular, if rarely in stock), but more available and affordable are the bare elements (available from Sparkfun) and industrial units that occasionally show up on the second-hand market. A few years ago I found a collection of used Sercel SG-10s being sold on Ebay by a mining company in Texas for an extremely reasonable price, and I thought I would take a crack at modifying them for audio use. It turns out that this is very easy to do! The Sercel units come wired together in a single chain (several of these chains are often used together in a grid pattern along with a giant vibrating truck to cover a large area when searching for oil reservoirs) with 4 signal wires. Each element uses two wires (I am not certain what the other two are for), which could be traced and soldered to standard XLR connectors. After testing, it became clear that the units were functional and outputting enough signal to be used with standard audio recorder preamps. Hurray!

A few observations about the Sercels after using them for some time:

The sensor enclosures and cables are designed for use in harsh industrial environments, and thus are extremely rugged and hardwearing. However, as a result they are also a bit bulky for regular carry usage, and I generally only bring them out into the field for specific projects. Additionally, the original cables are not shielded, meaning that in urban areas or elevated locations they are prone to picking up radio interference or 60-cycle hum from power lines. This has proven to be the most pervasive issue, and I plan to replace the cables with standard shielded microphone cable in the near future. They also lack some of the attachment options of commercial alternatives like the LOM Geofon; the stake appears to be permanently mounted to the enclosure, precluding the addition of a suction cup or magnetic attachments.

All in all though I am very impressed by the SG-10s, and I find them excellent for recording deep LFE rumble or for interesting pitched-up infrasound recordings.

Accelerometers

I was first introduced to industrial accelerometers by sound artist Dann Disciglio, who has used them in his excellent recordings of water flow in trees, among other things. These are testing units, usually used for quality control by manufacturers of vehicles and other industrial equipment. They are extremely sensitive, much more so than traditional piezo or geophone sensors, and thus unlock a whole new world of microscopic sound. However, they can be difficult to come by and are often prohibitively expensive. Two things are needed to use these units for audio recording purposes: the accelerometer unit itself and a separate power source/amplifier unit. Most of the units available on Ebay or elsewhere are made by Dytran or PCB Electronics, and the two are not (to my knowledge at least) cross-compatible. One must find a sensor and power source that will work well together, which can be a tricky task given the somewhat impenetrable data given by the manufacturers. The first units I purchased (PCB V353B13 and 480D06) required some modification to work: the sensor came with two stranded leads and no connector which I replaced with shielded audio cable, and the power source featured mini-BNC connectors that I replaced with 3.5mm jacks for ease of use. I have been testing this system since and it does produce very sensitive recordings, but it does still occasionally pick up radio interference at high gain settings; I will try adding more shielding to the power source to mitigate this issue.